Categories for Bootstrapped

I Want To Write About Your Startup – Relaunching TSF

tsfbig

I want to write about your startup. I’ve been putting together new content for the TSF relaunch and I wanted to reach out to longtime readers to see what you’ve been working on as a “thank you” for your support.

I’m relaunching TSF within the week with some killer content (Alexis Ohanian – cofounder of Reddit, Randall Bennett – On how his startup got on CNN, etc.) and would love to find a way to get your startup story in the mix.

Drop me an email – paul@thestartupfoundry.com and let’s talk about your startup.


Hopelessly perfect: Why it’s smart to work at a no-shot startup


Think about the promise of the Valley. Rampant hiring. Early-stage capital freely flowing to countless companies. Catered lunches and dinners. Worthy engineers treated like royalty (or worse, ninja / pirate / rockstars).

While some companies do fit the Valley pipe-dream, we think there are many more startups who don’t have a shot at VCs, the top engineers in the world, or an ice cube’s chance in Hell at realizing those paper stock options. And guess what: We think it’s still a smart idea to work with them. We call them the no-shot startup, and it can be an invaluable experience, just as long as you know what to expect and when to quit.

Hopeful startups

The startup lifestyle happened a bit backward for me. I accidentally worked for a successful and notable startup first, then started loving the startup lifestyle a few years later.

👉 Read Article
.

The Transparent Startup Experiment: A weekly look behind the scenes of “Time Off”


“I love reading stories about startups raising money, launching, or being acquired, but I want to learn how to take my startup to the next level with marketing. Is there any good examples I could check out? Where do I even start?”

I received this email from a TSF reader a few weeks ago, and I’ve been trying to figure out the best resource to send him. I sent him Noah Kagan’s Marketing plan for Mint (which is required reading), but I didn’t feel like it was enough. I wanted to be able to dissect a startup from top to bottom and show actual numbers as well as the thinking behind each decision. Beyond that, I wanted something that would teach other startups how to have the same success.

A finished product, with no marketing

A few months went by, but I still hadn’t found a resource like I described above. I was about to give up on the idea until I met Ryan Smith. Ryan is a very smart guy who loves building products. In fact, he built Time Off, an application that let’s small businesses employees easily manage vacation days, as a side project. It integrates with Google Apps and is a solid product. The problem is, Ryan wasn’t sure how to market it. I asked Ryan if I could come on board with him to do the marketing, and use Time Off as a real world case study for TSF.

We’re going to build a transparent startup that will walk TSF readers through a basic marketing plan, and share our results. This project will be a fantastic real world example to discover and learn what tactics actually work for startups. I’m very excited to bring this to TSF readers via a weekly column. It will be a fantastic resource to learn from, and I’ll do my best to explain the thinking behind our decisions. Be sure you follow TSF on Twitter @startupfoundry and on Facebook to stay current with “Transparent Startup Experiment”.

To get the ball rolling, this is the game plan that we’re going to be executing starting next week:

Our Game Plan For Time Off:

Objectives
a) $2,000 a month (in revenue) in 6 months.
b) Write up in 2 major publications

Target Users
1) Small Business
2) Google Apps for your domain users

Metrics
a) Page views
b) Number of users who sign up for trial that we convert – Who’s paying for our product (identify industries to target).
d) Measure the impact of weekly articles about Time Off on The Startup Foundry.

Registration Process
a) It’s already fast (which is really good), but it needs to be polished
b) Look at implementing Chargify instead of Google Checkout.
c) Use 37signals model of pricing.

Branding
a) Logo
b) Home page redesign
c) Improve copy on home page
d) The sales page needs to be redesigned.
e) Email through @timeoffhq.com instead of @gmail.
f) Social Proof needs a higher visual hierarchy on home and sales page

Increasing paid conversions
a) 1 month trial instead of free
b) The visual hierarchy of the site pushes customers to the free version. Let’s change this.
c) Collect emails, and have regular content go out to paid users

Blank State
a) Improve “first run” experience after signup.

What do you think? As you look at the site, do you see anything we missed?

Why Carbonmade decided to bootstrap and go freemium

When I was studying design a few years ago, I discovered many of my peers were struggling to get a portfolio online. Many of them didn’t have the required technical knowledge to properly showcase their work. Some of them hired friends in the Computer Science division but their portfolios would grow stale because they didn’t know how to update them.

Carbonmade is attempting to solve that problem by taking care of the technical details while letting artists focus on their work. I had the chance to catch up with a co-founder of Carbonmade, Spencer Fry, to talk about the early days. In this interview we cover how Carbonmade identified a problem, how they determined pricing, and why the “freemium” model is working for them.

Were you bootstrapped or funded?

We bootstrapped Carbonmade because it’s all we know. I’d started a company previously called TypeFrag. Bootstrapped that as well. For me, it just makes sense to collect revenue from the start of your business and if you can’t collect enough of it then you shouldn’t be in business. It’s just a natural progression. Can’t sell enough of what you’re trying to sell? Then you’ve got to shutdown and try something else.

There are companies that can’t be built this way such as Facebook, Tumblr or Twitter where you need a large engaged group of users before you can make any money. To bootstrap you typically need a single user to be able to benefit from the service without having to connect with anyone else. That’s Carbonmade. A user can sign up and create their own online portfolio without needing other people in the system to benefit from it.

We get plenty of offers from investors to talk, but we’re doing well and not having a $5 million check has never held us back from doing what we want. If money is ever the thing holding us back then we might re-consider taking outside investment.

What problem are you solving?

Carbonmade solves a really basic problem. There are millions of creative people in the world, but many of them either don’t have the skills required to design their own online portfolio or the time. Using Carbonmade means they spend less time fussing around with their portfolio and more time focused on developing their artwork.

What’s the biggest challenge you’ve overcome while building Carbonmade?

The biggest challenge for us and for all companies for the most part is hiring. We’re very selective in who we hire. We’re looking for specific individuals that are self-starters, independent workers that don’t need constant handholding, and have built or designed web applications before.

We hired our first people in August 2010. Up until then the three of us worked on Carbonmade for nearly four years. We’re now eight and excited to continue to build out the nucleus of our team.

How did you identify that there was a market for an online portfolio?

We stumbled upon the online portfolio market. Dave and Jason created the first version of Carbonmade back in December 2005 as a tool for Dave to update his own portfolio. He was tired of having to manually update his portfolio every time he created something new. Together they built it ‒ without a sign up page at first ‒ and Dave’s friends bugged him to open it up for them.

I came on about a year later, but we still weren’t working on it full-time. We were consulting and working on other apps, but Carbonmade kept slowly gaining traction. Starting around early 2010 we were able to stop taking on new client jobs and focus our time and energy on developing and supporting Carbonmade. That’s when Carbonmade really started to blow up.

How did you determine pricing?

We experimented with having two pricing points early on: $9 and $15. We found that nearly everyone was choosing the $9 plan and the $15 plan was being neglected. There weren’t enough features to warrant having two paid plans, so we settled on $12, because we thought that $9 was too low.

The question you want to ask yourself is: “What would I pay for this?” $12 felt right to us at the time. The product has improved significantly since then, which makes us happy that the value of our product has risen, but the price has remained the same.

Have you had success with your freemium model?

People generally ask me about our freemium business model. I believe that giving away a taste for free is a great marketing strategy. With Carbonmade, the free model works for many users in the beginning but as their career develops and they start getting more work, they need to upgrade to take advantage of our paid plan features. We’ve had dozens of people upgrade after 1,000 days which is a significant figure. Forcing them to pay on day one may have kept them from signing up in the first place.

What’s difficult about having a freemium business model is determining where to set the restrictions. The free plan needs to be good enough that it doesn’t feel crippled, but not so good they see no reason to upgrade to the paid plan.

If you need to showcase your work, be sure to checkout Carbonmade.

For more startup news, follow us on Twitter @startupfoundry or on Facebook.

How Jared Tame bootstrapped his startup by writing “Startups Open Sourced”

As Jared Tame was finishing up school, he decided to build a startup. He didn’t want to raise money (at least in the early stages) so he did what any “Grade-A” hustler would have done. Jared wrote a fantastic book on popular startups, Startups Open Sourced, and invested the profits into his idea.

In this interview Jared talks about where inspiration came from, measuring advertising efforts, and why Twitter isn’t an effective medium for advertising. This interview is worth investing 15 minutes into.

Final Thoughts:

Jared’s story reminds me of a previous post, You’re not entitled to anything. Hustle for everything you’re worth. Jared perfectly embodies this attitude.

For more startup news, follow us on Twitter @startupfoundry or on Facebook.

How a solo, non-technical founder hustled her way to launching a startup



As a solo founder, what’s been the biggest challenge you’ve had to overcome?

In a nutshell, you don’t know what you don’t know. You have to be more of a generalist as a sole founder vs. being able to really deep dive into any one expertise. So there is always a bit of unknown with regard to making sure you’ve covered everything. The best way to work around this is to meet with experts and mentors in areas you don’t know a lot about and ask them to grill you with questions about what your doing in that area of expertise. For me that meant meeting with several engineers, asking about development, database structure, hosting, security, coding etc. and finding out how to ask the right questions when finding an outsource firm. Whenever I get stuck now, I continue to reach out to my mentors and experts to ask for quick feedback or to ask the simple question, “What am I missing here?”

Tell us about your experience outsourcing development.

In the past I’ve had great luck outsourcing to sites like: Elance, Odesk, RentaCoder (now vWorker.com) etc., for coding projects. All of those projects, however, were very simple, non-complex type of development projects and outsourcing to these resources worked great. When it came to Swayable, I’d built a massive spec document, did a viability survey on Amazon’s Mechanical Turk and got to my minimum viable product spec so I was ready to go. I initially used Odesk and interviewed 5 firms as I wanted a company vs. an individual so I had a larger support team for the project as it grows. I found a firm that seemed to be able to complete the project but after a month past the first deadline, milestone 1 wasn’t met. I paid them for the work they did complete and ended the relationship. I then decided I needed a U.S. Based firm with U.S. Based accountability and then development could be done oversees. I hired Beyondsoft Consulting and they have been phenomenal! They’ve built both my iPhone and Web app as well as provided not only support, but they do a great job of questioning the direction at times and offering up better ways to execute if there are more efficient ways than what I’ve suggested. I couldn’t be happier with an outsource firm and will continue working with them for the foreseeable future. Lesson learned from this is really just jump in, try things out and don’t be afraid to move on to a better fit, after all it’s your product and money.

If you could do it all over again, what would you have done differently?

I was a bit nervous up front to alienate any audience as Swayable can be used in such a variety of ways. So I built the site with more gender neutral in mind, had I done some more surveys about customer segments, I would have learned that women and teens are more of my audience and could have built the website a bit more targeted to them. Without knowing the direction your site will take up front, and who will adopt your product the most (not all products have this problem if it’s a clear niche) it’s tough to figure out the right design and UX components. So looking back, I would have likely done some different branding elements and more gender targeting with design. It’s something I can fix in the future, but am wanting to continue to see how Swayable is adopted before making any drastic changes.

How do you plan to monetize?

With Swayable’s website only 3 months old and the iPhone app a couple of weeks old, some of the monetization options depend on which direction the site goes with the user base. I do have 4 monetization options regardless of direction that I’m working on implementing at certain traffic milestones:
1 – Paid API: Currently when 3rd party sites embed a Swayable I am seeing on average 10%+ click through/engagement with the Swayables. So an API is a natural solution for websites and online businesses & retailers that want to increase user engagement through there site and products and allowing users to create & share Swayable’s directly from there product/service pages without going to Swayable.com
2 – Selling Swayable advertisement spots – Soon, advertisers can purchase Swayables to get a more integrated/interactive advertising campaigns where users actually interact with the brand by “Swaying” on a Swayable.
3 – Viglink.com – Affiliate program, already installed on the site.
4 – Banner Advertising – Traditional banner advertising on the site.

How have you gained traction

If you really target an audience and learn how to connect with them, you should see traction (this does not mean you have to spend money to do this). In my opinion, If you don’t start seeing the traction you want, then you should be questioning either your product or pivot your product to see where you get traction. With Swayable, I have 3 primary audiences I am targeting initially and one of those audiences is teens (13 to early Twenties is how I categorize this). I’ve targeted them via content that they consume regularly online i.e. Celebrities fan sites via Facebook fan pages, blogs and twitter content that target these celebrities. I create content the teens want to consume and use Swayable as the medium in which they consume the celebrity content. As the teens visit Swayable to vote on celebrity content, they get engaged by first creating yet another celebrity style Swayable. Once the start seeing others vote on their Swayable and the addictiveness of seeing feedback real time, they start to create content in other categories and topics, which is exactly what I was hoping for. So far in 3 months I am doubling in visitors each month, tripling in page views, and increasing time on site each month. Best of all, users that don’t care about celebrity content can just filter by topic to not have to see any celebrity Swayables. My users are engaged, creating content, and over 50% of my visitors are returning back to the site to create and consume more. I am also seeing quite a bit of traction with website owners who are embedding Swayable’s right on their site, they are seeing on average 10% click through on the Swayable’s, getting more user engagement and interactivity without sending visitors away. Once a website owner has embedded a Swayable and seen the engagement, they are coming back to create and embed more Swayable’s on there site.

For more startup news, follow us on Twitter @startupfoundry or on Facebook.

-->

Lindsey Harper is a solo founder who doesn’t code. In all essence she is the antithesis of Valley culture. Yet this hasn’t stopped her from building a startup, Swayable, that’s gaining teen users hand over fist. I had the chance to catch up with Lindsey via email to talk about the challenges that she’s faced and how she overcame them.

What is Swayable?

Swayable helps you get opinions online quickly. For example:


As a solo founder, what’s been the biggest challenge you’ve had to overcome?

In a nutshell, you don’t know what you don’t know. You have to be more of a generalist as a sole founder vs. being able to really deep dive into any one expertise. So there is always a bit of unknown with regard to making sure you’ve covered everything. The best way to work around this is to meet with experts and mentors in areas you don’t know a lot about and ask them to grill you with questions about what your doing in that area of expertise. For me that meant meeting with several engineers, asking about development, database structure, hosting, security, coding etc. and finding out how to ask the right questions when finding an outsource firm. Whenever I get stuck now, I continue to reach out to my mentors and experts to ask for quick feedback or to ask the simple question, “What am I missing here?”

Tell us about your experience outsourcing development.

In the past I’ve had great luck outsourcing to sites like: Elance, Odesk, RentaCoder (now vWorker.com) etc., for coding projects. All of those projects, however, were very simple, non-complex type of development projects and outsourcing to these resources worked great. When it came to Swayable, I’d built a massive spec document, did a viability survey on Amazon’s Mechanical Turk and got to my minimum viable product spec so I was ready to go. I initially used Odesk and interviewed 5 firms as I wanted a company vs. an individual so I had a larger support team for the project as it grows. I found a firm that seemed to be able to complete the project but after a month past the first deadline, milestone 1 wasn’t met. I paid them for the work they did complete and ended the relationship. I then decided I needed a U.S. Based firm with U.S. Based accountability and then development could be done oversees. I hired Beyondsoft Consulting and they have been phenomenal! They’ve built both my iPhone and Web app as well as provided not only support, but they do a great job of questioning the direction at times and offering up better ways to execute if there are more efficient ways than what I’ve suggested. I couldn’t be happier with an outsource firm and will continue working with them for the foreseeable future. Lesson learned from this is really just jump in, try things out and don’t be afraid to move on to a better fit, after all it’s your product and money.

If you could do it all over again, what would you have done differently?

I was a bit nervous up front to alienate any audience as Swayable can be used in such a variety of ways. So I built the site with more gender neutral in mind, had I done some more surveys about customer segments, I would have learned that women and teens are more of my audience and could have built the website a bit more targeted to them. Without knowing the direction your site will take up front, and who will adopt your product the most (not all products have this problem if it’s a clear niche) it’s tough to figure out the right design and UX components. So looking back, I would have likely done some different branding elements and more gender targeting with design. It’s something I can fix in the future, but am wanting to continue to see how Swayable is adopted before making any drastic changes.

How do you plan to monetize?

With Swayable’s website only 3 months old and the iPhone app a couple of weeks old, some of the monetization options depend on which direction the site goes with the user base. I do have 4 monetization options regardless of direction that I’m working on implementing at certain traffic milestones:
1 – Paid API: Currently when 3rd party sites embed a Swayable I am seeing on average 10%+ click through/engagement with the Swayables. So an API is a natural solution for websites and online businesses & retailers that want to increase user engagement through there site and products and allowing users to create & share Swayable’s directly from there product/service pages without going to Swayable.com
2 – Selling Swayable advertisement spots – Soon, advertisers can purchase Swayables to get a more integrated/interactive advertising campaigns where users actually interact with the brand by “Swaying” on a Swayable.
3 – Viglink.com – Affiliate program, already installed on the site.
4 – Banner Advertising – Traditional banner advertising on the site.

How have you gained traction

If you really target an audience and learn how to connect with them, you should see traction (this does not mean you have to spend money to do this). In my opinion, If you don’t start seeing the traction you want, then you should be questioning either your product or pivot your product to see where you get traction. With Swayable, I have 3 primary audiences I am targeting initially and one of those audiences is teens (13 to early Twenties is how I categorize this). I’ve targeted them via content that they consume regularly online i.e. Celebrities fan sites via Facebook fan pages, blogs and twitter content that target these celebrities. I create content the teens want to consume and use Swayable as the medium in which they consume the celebrity content. As the teens visit Swayable to vote on celebrity content, they get engaged by first creating yet another celebrity style Swayable. Once the start seeing others vote on their Swayable and the addictiveness of seeing feedback real time, they start to create content in other categories and topics, which is exactly what I was hoping for. So far in 3 months I am doubling in visitors each month, tripling in page views, and increasing time on site each month. Best of all, users that don’t care about celebrity content can just filter by topic to not have to see any celebrity Swayables. My users are engaged, creating content, and over 50% of my visitors are returning back to the site to create and consume more. I am also seeing quite a bit of traction with website owners who are embedding Swayable’s right on their site, they are seeing on average 10% click through on the Swayable’s, getting more user engagement and interactivity without sending visitors away. Once a website owner has embedded a Swayable and seen the engagement, they are coming back to create and embed more Swayable’s on there site.

For more startup news, follow us on Twitter @startupfoundry or on Facebook.

Everyone has problems. Your solutions make you special. – Bootstrapping Postmark


I recently had the chance to sit down with Alex Hillman and Natalie Nagele from Postmark to give TSF readers an inside peak at this great transactional email company. In this interview we talked about what Postmark does, how they bootstrapped, Amazon entering their market, and customer acquisition. After reading this interview, be sure to checkout their website at www.PostmarkApp.com.

In one sentence, what does Postmark do?
Postmark helps other developers send important emails reliably.

What would be an example of an “important email”?
Marketing emails are valuable if they convert, but transactional emails are more valuable because they’re for your actual customers, not just random strangers.

If a marketing email doesn’t show up, it sucks. If a transactional email like a software license, receipt, invoice or some other important documentation doesn’t show up, you’re in trouble. We specialize in transactional email and, in fact, that’s the only thing we do with Postmark.

Were you bootstrapped or funded?
We are 100% bootstrapped. Wildbit has been doing design and development consulting since 1999. We built a R&D bucket and funded Beanstalk. Beanstalk grew and became the first product that had full time team members working on it. Beanstalk has been growing strong since ’07 and it gave us some runway (along with our existing knowledge and experience) to fire up Postmark in early ’10.

Amazon recently entered the transactional email business. How did Postmark react?
That was kind of a funny week. They had just released Elastic Beanstalk, a service totally different from OUR beanstalk app, but it had very similar branding. Then, a week later when they released an email delivery service we were like, “SERIOUSLY!?!?” Even their branding looked similar. It was pretty wild. Anyway, once we got past the initial OMG, it actually felt really good for three reasons.

1. Amazon validated the market. Anyone who hadn’t been thinking about looking for an outsourced email delivery host was now considering it.

2. We were validated as their competition. This was rewarding. It would have sucked if they came out and nobody thought of us. When, however, Amazon came out with their service, a common question was; “are Postmark and Sendgrid doomed?” We were actually validated as top-tier competition. This was a huge morale win.

3. Amazon won the race to the bottom (in terms of price). It no longer made sense to compete to be the cheapest, as they’d win by a long shot. The team talked and we agreed that if we weren’t going to compete on price we’d compete on reliability, support, ease of use, and, of course, our flare for style.

Essentially, Amazon showed up and offered well whiskey. We want to be top shelf Bourbon. We do a TON of hands-on work with our customers monitoring and diagnosing delivery systems.

How has your acquisition strategy evolved?
These days customer acquisitions are a bit different than when we first started Postmark. When Postmark started not many people thought about using a service instead of their mail server. The issue wasn’t, “why do I need Postmark?.” They weren’t even generally aware that a service like ours existed.

In the last year that’s changed quite a bit. The competitive landscape has widened and Amazon showed up a few months back to kindly validate our market.

Do you have any advice for other entrepreneurs?

1. Focus on one thing:

Postmark is the only provider that is solely transactional email. We don’t allow bulk marketing email at all.

2. JFDI (Just F’ing Do It):
That’s not “do EVERYTHING”, mind you but at a certain point, you’re sure to be over thinking it. JFDI = no excuses. Your problems aren’t what make you special. Everyone has problems. Your solutions make you special.

For more startup news, follow us on Twitter @startupfoundry.

From Freemium to Paid, to the cover of Hacker Monthly. The Bidsketch story.

I met Ruben Gamez in Miami at the SuperConf after-party. The day before the party, Ruben gave his 2 week notice at his “real job” so he could work full time on his startup, Bidsketch.

Bidsketch a web based proposal app for designers. Most designers write proposals using MS Word believe it or not. It’s completely old school and backwards. Bidsketch brings all this online and lets them do things that weren’t possible for them before. Rubin’s main vision is to change the way designers present and sell their services. So far, he’s happy with the results. Ruben agreed to an email interview and the transcript is below.

Paul Hontz: Ruben, talk to us about launching Bidsketch.

Ruben Gamez: I launched Bidsketch almost a year and a half ago. I had a day job so I worked nights and weekends to launch. It was a decent launch considering I was bootstrapping and had no connections or audience. I had a pre-launch to my mailing list and had over 20 paid accounts before it was officially live. In beta I also somehow managed to acquire (somehow, because I didn’t have money) my main and only competitor at the time (long story there and probably off topic). Editors Note: We’re going to bring you back and have you tell this story.


Paul: You initially launched as a freemium service. How did that work out for you?

Ruben:
After I launched I discovered that the freemium model was a major failure for me. After weeks of pulling my hair out and trying everything I could think of, I experimented my way out of the freemium model and have seen excellent growth since. I actually did a guest post on Rob Walling’s blog about that: Why Free Plans Don’t Work (It also made the cover of Hacker Monthly which was pretty exciting for me)

Anyways, over the next few months I got a fair amount of buzz (plus writeups) from the designer community and saw some nice growth. Funded competitors and teams came into the space which seems to have been a good thing because I’m growing faster since then and I outrank them in Google which brings in great traffic.

Paul: What strategies have you used for user acquisition?

Ruben: My strategy has been a combination of word of mouth, relevant writeups, SEO, and some content marketing (I don’t advertise).

At this point it’s become profitable enough to cover all my living expenses (which are pretty real expenses considering that I’m married and have a mortgage). A couple of weeks ago I quit the day job; So now I’m officially 100% full time on Bidsketch.

Paul: What’s next for Bidsketch? How are you doing on revenue?

Ruben: I’m very excited about what’s next. Aside from being able to devote some real time to working on what I love, in a month or so I’ll be releasing version 2 of Bidsketch which has some pretty neat stuff that’s going to knock the socks off designers.

Our monthly recurring revenue has more than doubled in the last four months.

Paul: Do you have any advice for solo-founders?

Ruben: Just giving an overview like this makes it seem like it’s been a walk in the park but the truth is that it’s been hard. It’s hard work consistently grinding it out after-hours for over a year as a single founder. Along with many of the highs there were definitely some rough times. Hard work but totally worth it — especially now that I’ve come through the other side.

To checkout Ruben’s company, check out Bidsketch.com. For up to the minute startup news, follow us on Twitter @startupfoundry.

31 days, 120,000 hits, and $462 in revenue. The Startup Foundry’s story.


In our first month online The Startup Foundry had 120,000 page views, grew to over a thousand followers on twitter, and generated $462 in advertising revenue. I thought I would share the backstory and some lessons we learned along the way in hopes that it might be useful for other startups.

The Back Story.

The idea for TSF came to me while I was building my other startup, CodeSketch. I love talking and writing about startups but I live in West Michigan and the startup scene is rather small. I wanted to connect with other entrepreneurs that were building really cool things across the world and I saw The Startup Foundry as a vehicle to make that happen. TSF is a completely moonlit project.

Lessons learned:

1. Once you have identified a need, DO SOMETHING immediately.
•Don’t wait for other people to help you. Jump in and do it yourself. You can end up spending more time trying to convince other people to join you then just doing the work yourself.

•If you’re waiting to have everything completely polished, you’ve waited to long.

•Be borderline obsessed. I’m so incredibly hungry to make TSF the best place for entrepreneurs and founders on the web. Don’t lose focus.

2. Be the first to admit you suck at something, then work your ass off to fix it.
•When we first launched the site the comments were easier to read then the articles. Every startup is going to make stupid mistakes that seem blatantly obvious in retrospect. Don’t let this scare you.

•Our videos interviews were terribly produced (our guests had fantastic content though). If you watch my first interview that I published with Andrew Warner, you almost feel sorry for me I did such a poor job. I’m still not quite to the level I want to be at yet but I’m working at it and I’m getting better.

•Your goal is to not suck twice.

3. Be proud of your accomplishments, but not cocky.
•In the grand scheme of things we are a tiny site, but this is a tiny site I’m incredibly proud of. I love waking up each morning to an inbox stuffed with fantastic startup news that I have the privilege to dig through. It is truly humbling to have guys like Andrew Warner, Tim O’Reilly, and Alexis Ohanian give you a half hour of their time. I love that I have the privilege to do this and I don’t take it for granted.

•I’ve talked to some entrepreneurs that are almost afraid to take credit for their success. Luck is an admittedly huge role in success but take credit for showing up to work everyday. Many people fizzle out in the day to day grind. Take pride in your accomplishments (even if your accomplishment is that you’re still moving an inch at a time).

The Future.

After seeing the initial traction, I’m going to start shifting more of my time to The Startup Foundry and try to turn it into a powerhouse of startup news. I would love to see TSF grow into a full time job (complete with a full time salary), but we’re not there quite yet. I’m sure we are going to make more mistakes down the road but I promise that we will iterate quickly, and try to serve this community as best as we can. If you have any questions for me, feel free to ask them in the comments. I’ll try to answer them all. If you have an awesome startup, I want to hear about it, so send us a tip!

For more startup news, follow us on Twitter @startupfoundry.

O’Reilly: When something is commoditized, an adjacent thing becomes valuable

Summary of the video:
1. In this video Tim talks about what “Web 2.0” means. (How he views the internet as an operating system).

2. Tim realized that software was becoming a commodity. He saw the same thing play out earlier with hardware in the IBM era. When something is commoditized, an adjacent thing becomes valuable. In the IBM era when hardware was commoditized, software became very valuable (and Microsoft won).

3. The web is a platform. The example Tim gave was Napster. Napster worked by indexing the songs you already had on your computer, not from a pre-determined master list. This was a huge shift in thinking about what the web could be about.

4. Tim encourages entrepreneurs to not “paint by numbers” when it comes to solving problems. Just because one company was successful doesn’t mean you should clone their product.

Be sure to checkout part 1 of this interview where Tim talks about how he Bootstrapped O’Reilly Media with “happy accidents”

For more startup news, follow us on Twitter @startupfoundry.